Generalist Vision Foundation Models for Medical Imaging: A Case Study of Segment Anything Model on Zero-Shot Medical Segmentation

CauDR: A causality-inspired domain generalization framework for fundus-based diabetic retinopathy grading

Diabetic retinopathy (DR) is the most common diabetic complication, which usually leads to retinal damage, vision loss, and even blindness. A computer-aided DR grading system has a significant impact on helping ophthalmologists with rapid screening and diagnosis. Recent advances in fundus photography have precipitated the development of novel retinal imaging cameras and their subsequent implementation in clinical practice. However, most deep learning-based algorithms for DR grading demonstrate limited generalization across domains. This inferior performance stems from variance in imaging protocols and devices inducing domain shifts. We posit that declining model performance between domains arises from learning spurious correlations in the data. Incorporating do-operations from causality analysis into model architectures may mitigate this issue and improve generalizability. Specifically, a novel universal structural causal model (SCM) was proposed to analyze spurious correlations in fundus imaging. Building on this, a causality-inspired diabetic retinopathy grading framework named CauDR was developed to eliminate spurious correlations and achieve more generalizable DR diagnostics. Furthermore, existing datasets were reorganized into 4DR benchmark for DG scenario. Results demonstrate the effectiveness and the state-of-the-art (SOTA) performance of CauDR. Diabetic retinopathy (DR) is the most common diabetic complication, which usually leads to retinal damage, vision loss, and even blindness. A computer-aided DR grading system has a significant impact on helping ophthalmologists with rapid screening and diagnosis. Recent advances in fundus photography have precipitated the development of novel retinal imaging cameras and their subsequent implementation in clinical practice. However, most deep learning-based algorithms for DR grading demonstrate limited generalization across domains. This inferior performance stems from variance in imaging protocols and devices inducing domain shifts. We posit that declining model performance between domains arises from learning spurious correlations in the data. Incorporating do-operations from causality analysis into model architectures may mitigate this issue and improve generalizability. Specifically, a novel universal structural causal model (SCM) was proposed to analyze spurious correlations in fundus imaging. Building on this, a causality-inspired diabetic retinopathy grading framework named CauDR was developed to eliminate spurious correlations and achieve more generalizable DR diagnostics. Furthermore, existing datasets were reorganized into 4DR benchmark for DG scenario. Results demonstrate the effectiveness and the state-of-the-art (SOTA) performance of CauDR.

A Foundation Model for Cell Segmentation

Cells are a fundamental unit of biological organization, and identifying them in imaging data - cell segmentation - is a critical task for various cellular imaging experiments. While deep learning methods have led to substantial progress on this problem, most models in use are specialist models that work well for specific domains. Methods that have learned the general notion of "what is a cell" and can identify them across different domains of cellular imaging data have proven elusive. In this work, we present CellSAM, a foundation model for cell segmentation that generalizes across diverse cellular imaging data. CellSAM builds on top of the Segment Anything Model (SAM) by developing a prompt engineering approach for mask generation. We train an object detector, CellFinder, to automatically detect cells and prompt SAM to generate segmentations. We show that this approach allows a single model to achieve human-level performance for segmenting images of mammalian cells (in tissues and cell culture), yeast, and bacteria collected across various imaging modalities. We show that CellSAM has strong zero-shot performance and can be improved with a few examples via few-shot learning. We also show that CellSAM can unify bioimaging analysis workflows such as spatial transcriptomics and cell tracking. A deployed version of CellSAM is available at https://cellsam.deepcell.org/.

Technical Note: Generalizable and Promptable Artificial Intelligence Model to Augment Clinical Delineation in Radiation Oncology

BACKGROUND

Efficient and accurate delineation of organs at risk (OARs) is a critical procedure for treatment planning and dose evaluation. Deep learning-based auto-segmentation of OARs has shown promising results and is increasingly being used in radiation therapy. However, existing deep learning-based auto-segmentation approaches face two challenges in clinical practice: generalizability and human-AI interaction. A generalizable and promptable auto-segmentation model, which segments OARs of multiple disease sites simultaneously and supports on-the-fly human-AI interaction, can significantly enhance the efficiency of radiation therapy treatment planning.

PURPOSE

Meta's segment anything model (SAM) was proposed as a generalizable and promptable model for next-generation natural image segmentation. We further evaluated the performance of SAM in radiotherapy segmentation.

METHODS

Computed tomography (CT) images of clinical cases from four disease sites at our institute were collected: prostate, lung, gastrointestinal, and head & neck. For each case, we selected the OARs important in radiotherapy treatment planning. We then compared both the Dice coefficients and Jaccard indices derived from three distinct methods: manual delineation (ground truth), automatic segmentation using SAM's 'segment anything' mode, and automatic segmentation using SAM's 'box prompt' mode that implements manual interaction via live prompts during segmentation.

RESULTS

Our results indicate that SAM's segment anything mode can achieve clinically acceptable segmentation results in most OARs with Dice scores higher than 0.7. SAM's box prompt mode further improves Dice scores by 0.1∼0.5. Similar results were observed for Jaccard indices. The results show that SAM performs better for prostate and lung, but worse for gastrointestinal and head & neck. When considering the size of organs and the distinctiveness of their boundaries, SAM shows better performance for large organs with distinct boundaries, such as lung and liver, and worse for smaller organs with less distinct boundaries, like parotid and cochlea.

CONCLUSIONS

Our results demonstrate SAM's robust generalizability with consistent accuracy in automatic segmentation for radiotherapy. Furthermore, the advanced box-prompt method enables the users to augment auto-segmentation interactively and dynamically, leading to patient-specific auto-segmentation in radiation therapy. SAM's generalizability across different disease sites and different modalities makes it feasible to develop a generic auto-segmentation model in radiotherapy.

MRI radiomics-based decision support tool for a personalized classification of cervical disc degeneration: a two-center study

Objectives: To develop and validate an MRI radiomics-based decision support tool for the automated grading of cervical disc degeneration. Methods: The retrospective study included 2,610 cervical disc samples of 435 patients from two hospitals. The cervical magnetic resonance imaging (MRI) analysis of patients confirmed cervical disc degeneration grades using the Pfirrmann grading system. A training set (1,830 samples of 305 patients) and an independent test set (780 samples of 130 patients) were divided for the construction and validation of the machine learning model, respectively. We provided a fine-tuned MedSAM model for automated cervical disc segmentation. Then, we extracted 924 radiomic features from each segmented disc in T1 and T2 MRI modalities. All features were processed and selected using minimum redundancy maximum relevance (mRMR) and multiple machine learning algorithms. Meanwhile, the radiomics models of various machine learning algorithms and MRI images were constructed and compared. Finally, the combined radiomics model was constructed in the training set and validated in the test set. Radiomic feature mapping was provided for auxiliary diagnosis. Results: Of the 2,610 cervical disc samples, 794 (30.4%) were classified as low grade and 1,816 (69.6%) were classified as high grade. The fine-tuned MedSAM model achieved good segmentation performance, with the mean Dice coefficient of 0.93. Higher-order texture features contributed to the dominant force in the diagnostic task (80%). Among various machine learning models, random forest performed better than the other algorithms (p < 0.01), and the T2 MRI radiomics model showed better results than T1 MRI in the diagnostic performance (p < 0.05). The final combined radiomics model had an area under the receiver operating characteristic curve (AUC) of 0.95, an accuracy of 89.51%, a precision of 87.07%, a recall of 98.83%, and an F1 score of 0.93 in the test set, which were all better than those of other models (p < 0.05). Conclusion: The radiomics-based decision support tool using T1 and T2 MRI modalities can be used for cervical disc degeneration grading, facilitating individualized management.

On the challenges and perspectives of foundation models for medical image analysis

This article discusses the opportunities, applications and future directions of large-scale pretrained models, i.e., foundation models, which promise to significantly improve the analysis of medical images. Medical foundation models have immense potential in solving a wide range of downstream tasks, as they can help to accelerate the development of accurate and robust models, reduce the dependence on large amounts of labeled data, preserve the privacy and confidentiality of patient data. Specifically, we illustrate the "spectrum" of medical foundation models, ranging from general imaging models, modality-specific models, to organ/task-specific models, and highlight their challenges, opportunities and applications. We also discuss how foundation models can be leveraged in downstream medical tasks to enhance the accuracy and efficiency of medical image analysis, leading to more precise diagnosis and treatment decisions.

Enhancing Microdroplet Image Analysis with Deep Learning

Microfluidics is a highly interdisciplinary field where the integration of deep-learning models has the potential to streamline processes and increase precision and reliability. This study investigates the use of deep-learning methods for the accurate detection and measurement of droplet diameters and the image restoration of low-resolution images. This study demonstrates that the Segment Anything Model (SAM) provides superior detection and reduced droplet diameter error measurement compared to the Circular Hough Transform, which is widely implemented and used in microfluidic imaging. SAM droplet detections prove to be more robust to image quality and microfluidic images with low contrast between the fluid phases. In addition, this work proves that a deep-learning super-resolution network MSRN-BAM can be trained on a dataset comprising of droplets in a flow-focusing microchannel to super-resolve images for scales ×2, ×4, ×6, ×8. Super-resolved images obtain comparable detection and segmentation results to those obtained using high-resolution images. Finally, the potential of deep learning in other computer vision tasks, such as denoising for microfluidic imaging, is shown. The results show that a DnCNN model can denoise effectively microfluidic images with additive Gaussian noise up to σ = 4. This study highlights the potential of employing deep-learning methods for the analysis of microfluidic images.

Unraveling the complexity of Optical Coherence Tomography image segmentation using machine and deep learning techniques: A review

Optical Coherence Tomography (OCT) is an emerging technology that provides three-dimensional images of the microanatomy of biological tissue in-vivo and at micrometer-scale resolution. OCT imaging has been widely used to diagnose and manage various medical diseases, such as macular degeneration, glaucoma, and coronary artery disease. Despite its wide range of applications, the segmentation of OCT images remains difficult due to the complexity of tissue structures and the presence of artifacts. In recent years, different approaches have been used for OCT image segmentation, such as intensity-based, region-based, and deep learning-based methods. This paper reviews the major advances in state-of-the-art OCT image segmentation techniques. It provides an overview of the advantages and limitations of each method and presents the most relevant research works related to OCT image segmentation. It also provides an overview of existing datasets and discusses potential clinical applications. Additionally, this review gives an in-depth analysis of machine learning and deep learning approaches for OCT image segmentation. It outlines challenges and opportunities for further research in this field.